Skip to main content

Full text of "Email Correspondence between Aaron Swartz and Brett Bonfield, January 27 and 28, 2008"

See other formats


Brett Bonfield, Jan 27, 2008 at 9:50 PM

I've attached a summary of the evaluations on your presentation. I was really pleased by the number of people who seem to have gotten it.

I've been thinking about the idea of corruption since you brought it up a couple of weeks ago. I've seen Lessig's interview on German (?) television on YouTube, read his "Required Reading: the next 10 years," heard his "alpha" lecture and read your response (and his response to your response), and scanned the wiki page on corruption. Which is meant as a caveat: I've got more research to do. But I have been giving it a lot of thought.

When it comes to corruption, the analogy that works for me is the one about hockey players that Surowiecki describes in "Fuel for Thought" (http://www.newyorker.com/talk/financial/2007/07/23/070723ta_talk_surowiecki). Most of the players want to wear helmets, but won't admit it publicly. Because there's no rule that they must wear helmets, few of them do.

Lessig's analogy about the people who saw the headmaster abusing people and did nothing fails, for me, for three reasons:

I believe most people are not only basically good but they are *highly motivated to do good*.
People are often behaving rationally even when they act against their own interest, including those times when they don't do good even though they'd like to.
Giving people the opportunity to report anonymously and without negative repercussion, either that they'd prefer that everyone wear a helmet while playing hockey or that the headmaster is a monster, shifts the incentives. It makes it more rational for them to do good than not to do good.

This, I think, is where technology can come into play.

What strikes me about the greatest act of corruption disclosure this country has ever seen, the Watergate investigation, is how important it was to Deep Throat to remain anonymous, even after his allegations were turned into a criminal case and the Watergate conspirators were revealed. If the film of "All the President's Men" is even partially true, it was extraordinarily difficult for this informer to make Woodward and Bernstein aware of the corruption without sacrificing his anonymity. I realize Lessig is talking more about "influence" than criminal acts, but I think the method for revealing either often involves similar barriers. People are scared to come forward. This is a problem that technology can play a large role in solving. Not easily, but we know enough about privacy that we can make it possible for people to reveal information without anyone having a way to trace the path of that information back to them. The nature of the information may reveal something about them, but through technology we can effectively obscure the path. So the first thing that would need to be built would be a mechanism for giving anonymous tips.

It would probably also make sense to let the person giving the tips choose precisely how much they want to reveal about themselves and the information they're disclosing and to whom, and perhaps even dictate how they want the information used. Something along the lines of Creative Commons' modules might work here. I'm not a lawyer, nor have been in Deep Throat's shoes. I'm not sure how I would feel about sharing the sort of information he disclosed, but I firmly believe that I'd be more likely to reveal it if it didn't require running around in a trenchcoat and sneaking into and out of parking garages. Removing barriers would make it more likely that I'd reveal what I knew.

So you build this mechanism and every lunatic in the world starts implicating everyone else. How to separate a few vital grains of wheat from an avalanche of chaff? A nice first run would likely take advantage of SpamBayes, etc. Perhaps something along the lines of Robot9000 (http://blag.xkcd.com/2008/01/14/robot9000-and-xkcd-signal-attacking-noise-in-chat/).
But that's just the first run. After that? I think something along the lines of Reddit would work beautifully. If you or I, provided we have sufficient karma, give a tip an upvote, that should mean something. And, perhaps, once a tip has been given enough points from enough people with enough karma, it automatically gets its own Wiki page where it can be reviewed and, ideally, corroborated or disproved.

Is this far-fetched? Probably. But everything about Lessig's project is pretty far-fetched. What I like about this idea is twofold.

First, it gives anyone a chance to take action. Know something damning? In five minutes you can get it off your chest. Don't know anything damning? You can get involved in either the vetting or the investigation process. So people have something to do, they feel empowered. Part of the wise crowd gets to play Deep Throat, the other part Woodward and Bernstein.

Second, it makes it much harder to keep secrets. How many people would have to be embarrassed (or prosecuted) based on corroborated stories before people get a whole lot more concerned about being found out and become a whole lot less corruptible? The first one or two victories, given sufficient media play (which they'd certainly get), could be inordinately influential.

Anyway, that's my pie in the sky. Ignore it or cut it to ribbons. But it's probably the best I have to offer.

--

Aaron Swartz, Jan 28, 2008 at 12:06 PM

That's an interesting take. I have some friends who are trying to
attack things thru this angle:

http://wikileaks.org/wiki/Wikileaks

There's some pretty interesting stuff on there already. They have some work to do on UI -- you could probably help a bunch if you got involved -- but I think they've got the right idea.

But I tend to disagree with the if-only-they-knew-the-truth school of thought. Watergate happened not because the story came out -- COINTELPRO started in 1956; stories like this came out all the time in the independent press -- it was because Nixon went after someone powerful (the DNC) who could fight back. Had it been Nixon burglarizing the Socialist Worker's Party offices again, the Post never would given the story such attention and Woodward and Bernstein would have been stayed on the cub beat. So airing the stories is good, but it's nowhere near enough. We need an alternate system for making them interesting and getting them to people. And that's much harder.

--

Aaron Swartz, Jan 28, 2008 at 12:07 PM

> but it's nowhere near enough. We need an alternate system for making
> them interesting and getting them to people. And that's much harder.

And even when you do that... what do people do in response to hearing there's all sorts of bad stuff going on? That's the part I'm interested in.

--

Brett Bonfield, Jan 28, 2008 at 12:49 PM

I think what we need are prosecutions. That's what interests me about leaks. Wikileaks looks interesting, but it's still just an alternative press. And, as you've pointed out (and Lessig points out), people ignore the alternative press, and even the scholarly journals and much of the mainstream media.

Let's say the election comes down to Clinton against Huckabee (pick any two politicians and any race, this is just an illustration). What if the Sunlight Foundation and Wikileaks and your project with Lessig present me with clear evidence that Clinton is on the take, that she's been funded by dozens of lobbies I distrust. OTOH, Huckabee's honest through and through, nothing but individual contributions from voters like me. Am I supposed to vote for Huckabee?

I don't see the connection between more information and a change in the system. I do see a connection between the right information and removing the corrupt from positions of power.

--

Aaron Swartz, Jan 28, 2008 at 12:53 PM

> I don't see the connection between more information and a change in
> the system. I do see a connection between the right information and
> removing the corrupt from positions of power.

What's the connection?

--

Brett Bonfield, Jan 28, 2008 at 12:57 PM

right information = sufficient data to justify prosecuting a case

removal from power = jail time (or Nixon-style resignation)

--

Aaron Swartz, Jan 28, 2008 at 1:02 PM

> > > I don't see the connection between more information and a change in
> > > the system. I do see a connection between the right information and
> > > removing the corrupt from positions of power.
> >
> > What's the connection?
>
> right information = sufficient data to justify prosecuting a case

Why don't prosecutors solicit leaks then? What pressures the prosecutors?

--

Brett Bonfield, Jan 28, 2008 at 1:22 PM

I'm sure they do solicit leaks, but I have yet to see anyone put forth the idea that prosecutors aren't understaffed and underfunded. I also believe, as I've mentioned earlier, that providing leaks is harder than it needs to be.

However, I see two pressures supporting prosecution.

1. Opposition. Most lobbies have opposing lobbies. Not all, but
there's not a lot of incentive to create a powerful lobby unless
there's someone to oppose. Most corporations have competitors, as do
most politicians. As you point out, the Democrats were powerful enough to help see to it that Watergate conspirators were prosecuted.

2. Money. The cigarette industry is still powerful, but it's a lot
less so after weathering so many lawsuits. While I favor tort reform,
even with reform it's likely going to be worth it for people to seek
damages from corrupt corporate executives, elected officials, etc.